Place conditioning with the dopamine D1-like receptor agonist SKF 82958 but not SKF 81297 or SKF 77434

European Journal of Pharmacology 343 Ž1998. 111–118

Place conditioning with the dopamine D 1-like receptor agonist SKF 82958 but not SKF 81297 or SKF 77434 Brett S. Abrahams a , Jeff D. Rutherford a , Paul E. Mallet a , Richard J. Beninger a b

a,b,)

Department of Psychology, Queen’s UniÕersity, Kingston, Ont., Canada K7L 3N6 Department of Psychiatry, Queen’s UniÕersity, Kingston, Ont., Canada K7L 3N6

Received 26 August 1997; revised 10 November 1997; accepted 18 November 1997

Abstract While self-administration and place conditioning studies have shown that dopamine D 2-like receptor agonists produce reward-related learning, the effects of dopamine D 1-like receptor agonists remain equivocal. The present study tested three dopamine D1-like receptor agonists for their ability to induce a place preference. Like control rats treated with amphetamine Ž2.0 mgrkg i.p.., rats treated with SKF 82958 Ž"-6-chloro-7,8-dihydroxy-3-allyl-1-phenyl-2,3,4,5-tetrahydro-1-phenyl-1H-3-benzazepine hydrobromide; 0.05 but not 0.01, 0.025, 0.075, or 0.10 mgrkg s.c. andror i.p.. during conditioning showed a significant increase in the amount of time spent on the drug-paired side during the drug free test. Neither SKF 81297 Ž"-6-chloro-7,8-dihydroxy-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrobromide; 0.25, 0.50, 1.0, 2.0, and 4.0 mgrkg i.p.. nor SKF 77434 Ž"-7,8-dihydroxy-3-allyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrochloride; 0.20, 1.0, 5.0, and 10.0 mgrkg i.p.. produced place conditioning. Significant increases in locomotion were seen at some doses of all drugs. Results show for the first time that systemic administration of a dopamine D 1-like receptor agonist produces a place preference and are consistent with previous findings showing that dopamine D1-like receptor activation produces reward-related learning. q 1998 Elsevier Science B.V. Keywords: SKF 82958; SKF 81297; SKF 77434; Dopamine; Dopamine D1-like receptor; Place conditioning; Learning, reward-related

1. Introduction A great deal of work implicates mesolimbic dopamine transmission in reward-related learning Žfor review, see Willner and Scheel-Kruger, 1991., but less is known about ¨ how different dopamine receptor subtypes mediate this learning. Prior to identification of the five currently recognized dopamine receptor subtypes ŽD 1 – 5 ., dopamine receptors were divided into two groups based on their effect on the enzyme adenylate cyclase. Activation of dopamine D1 receptors was shown to result in stimulation of adenylate cyclase, whereas dopamine D 2 receptor activation either inhibited or failed to affect this enzyme ŽKebabian and Calne, 1979.. Dopamine D 1 and D5 receptors have been found to be D 1-like but at present pharmacological agents acting selectively at either subtype are not available. Dopamine D 2 , D 3 and D4 receptors are D 2-like ŽSeeman and Van Tol, 1994.. ) Corresponding author. Tel.: q1-613-5452486; fax: q1-613-5452499; e-mail: [email protected]

0014-2999r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 0 1 4 - 2 9 9 9 Ž 9 7 . 0 1 5 3 1 - 8

The place conditioning paradigm has been used extensively to ascertain the role of dopamine D 1- and D 2-like receptors in reward-related learning Žsee Beninger, 1993; Beninger and Miller, 1997 for reviews.. Generally, results have supported a role for dopamine D 2-like receptors in reward, but corresponding data for dopamine D 1-like receptor involvement have been equivocal. Antagonism of either dopamine D 1- or D 2-like receptors abolished the rewarding properties of dopaminergic drugs in rats, suggesting that both families of dopamine receptors may be involved in reward-related learning ŽSpyraki et al., 1982; Mackey and van der Kooy, 1985; Mithani et al., 1986; Hoffman and Beninger, 1989; Hiroi and White, 1991.. In agreement with these results, dopamine D 2-like receptor agonists produced a place preference ŽMorency and Beninger, 1986; Hoffman et al., 1988; Hoffman and Beninger, 1988, 1989; White et al., 1991.. However, results with the dopamine D 1-like receptor agonist SKF 38393 were not straightforward. When injected into the nucleus accumbens, SKF 38393 produced a place preference ŽWhite et al., 1991., yet systemic administration

112

B.S. Abrahams et al.r European Journal of Pharmacology 343 (1998) 111–118

produced an aversion ŽHoffman and Beninger, 1988, 1989; White et al., 1991.. Perhaps the aversion was the result of pharmacological properties specific to SKF 38393. The possibility that pharmacological characteristics unique to SKF 38393 may be responsible for behavioral findings in place conditioning is supported by results from self-administration studies. In agreement with place conditioning results, pretreatment with either dopamine D 1- or D 2-like receptor antagonists disrupted self-administration of cocaine or amphetamine in rodents and primates, supporting the idea that both dopamine receptor families may be involved in reward ŽYokel and Wise, 1975, 1976; Risner and Jones, 1976; De Wit and Wise, 1977; Roberts and Vickers, 1984, 1987; Koob et al., 1987; Bergman et al., 1990; Corrigall and Coen, 1991; Caine and Koob, 1994.. Also similar to place preference studies, dopamine D 2-like receptor agonists were self-administered ŽWoolverton et al., 1984; Wise et al., 1990. and the dopamine D 1-like receptor agonist SKF 38393 was not ŽWoolverton et al., 1984; Weed and Woolverton, 1995.. However, other dopamine D 1-like receptor agonists from the same chemical family Žbenzazepines. recently have been shown to support self-administration: SKF 82958 in rats and squirrel monkeys, SKF 81297 in rhesus and squirrel monkeys, and SKF 77434 in rats but not squirrel monkeys ŽSelf and Stein, 1992; Self et al., 1993, 1996; Weed et al., 1993; Grech et al., 1996.. When all four benzazepine-based compounds were compared in a single study of rhesus monkeys, only SKF 82958 and SKF 81297 were self-administered; responding for SKF 38393 and SKF 77434 was not different from responding for vehicle ŽWeed and Woolverton, 1995.. Differences in pharmacological properties of the dopamine D 1-like receptor agonists may help to make sense of these results. Thus, SKF 82958 has been shown to be a full dopamine receptor agonist in both the rat and squirrel monkey, with observations of adenylate cyclase activation ranging from 78–148% of dopamine ŽO’Boyle et al., 1989; Izenwasser and Katz, 1993.. SKF 81297 has also been reported to be a high efficacy dopamine receptor agonist in the rat, with corresponding values of 68–88% ŽAnderson and Jansen, 1990; Arnt et al., 1992; Izenwasser and Katz, 1993.. In contrast, SKF 77434 and SKF 38393 are generally considered to be low efficacy agents, with adenylate cyclase responses of about 48–55% and 44–59% of dopamine, respectively ŽAnderson et al., 1985; O’Boyle et al., 1989; Izenwasser and Katz, 1993.. It has been suggested elsewhere that our understanding of the functional role of dopamine D 1-like receptors has been obscured by the use of SKF 38393 ŽSelf and Stein, 1992.. Thus, despite negative results with SKF 38393, recent results from self-administration studies suggest that activation of the dopamine D 1-like receptor indeed may be important to reward. It is hypothesized here that administration of high efficacy dopamine D 1-like receptor agonists, previously shown to be self-administered, will result

in a place preference. Three dopamine D 1-like receptor agonists ŽSKF 82958, SKF 81297, and SKF 77434. were administered systemically at a range of doses to test this hypothesis; amphetamine Ž2.0 mgrkg, i.p.. and distilled water Ži.p.. were also tested as paradigm controls.

2. Materials and methods 2.1. Subjects Two-hundred and nine male Wistar rats ŽCharles River, Canada., weighing between 200 and 250 g upon arrival at the colony had free access to both food and water in their home cages, and were group housed Žfour per cage. in a temperature controlled Ž218C. facility with a 12 h lightrdark cycle Žlights on at 08.00 h.. Care and treatment of the animals was in full compliance with guidelines set forth by the Canadian Council on Animal Care, the Animals for Research Act, and relevant Queen’s University policy. Rats were randomly assigned to treatment groups with approximately ten animals per group Žsee Fig. 1 or Section 2.4.. Only five rats were treated with 4.0 mgrkg SKF 81297 because of insufficient amount of drug. The number of animals in the amphetamine group Ž n s 20. was larger than other treatment groups as a result of an attempt to ensure paradigm reliability. Not all animals treated with amphetamine were tested at once. Instead, four animals were run at a given time along with animals from another group Že.g. 0.01 mgrkg SKF 82958 s.c... The large number of doses evaluated resulted in a relatively large number of animals in the amphetamine group. 2.2. Apparatus The test environment consisted of two separate chambers Ž38 = 27 = 34 cm. connected by a tunnel Ž8 = 8 = 8 cm.. Opaque guillotine doors could be used to block access to the tunnel. Six photocells Žtwo in each section of the apparatus. were connected to a computer that monitored animal position and locomotor activity. The left and right chambers had different walls Žeither sealed wood or plexiglass with alternating black and white vertical stripes. and floors Žeither parallel stainless steel bars or galvanized hardware cloth.. Four such units were employed; among the four units, each of the wall types appeared on both the left and right sides of the apparatus with each of the two floor types. Each unit was illuminated by its own 7.5 W bulb, and housed inside a styrofoam-insulated box to attenuate background noise. For more information see Brockwell et al. Ž1996.. 2.3. Procedure After arrival in the colony animals were handled daily for 6–8 days. The experimental protocol involved three

B.S. Abrahams et al.r European Journal of Pharmacology 343 (1998) 111–118

distinct phases: preconditioning, conditioning, and test, with each animal being tested once a day at approximately the same time each day during the light portion of the lightrdark cycle for fourteen consecutive days. 2.3.1. Preconditioning On days 1–3, animals were placed for 15 min in one of the two chambers with the tunnel open and then returned to their home cages. Start side was varied across animals, with half beginning on the right for all preconditioning days, and the other half on the left. The amount of time spent in each of the two chambers and tunnel was recorded to establish a baseline with which to compare test data. 2.3.2. Conditioning On days 4–11 rats were injected systemically with a drug or its vehicle, held in a metal transporting cage Žsee below., and then confined to one chamber of the place conditioning apparatus for 30 min. Drug was given on conditioning days 1, 3, 5, and 7 Žand always paired with one chamber. and the drug’s vehicle on remaining conditioning days Žand paired with the other chamber.. Total number of photocell beam breaks was recorded to assess locomotor activity. 2.3.3. Testing Days 12–14 were identical to preconditioning days. Animals were placed in the apparatus with the tunnel open for 15 min, then returned to their home cages. As before, half the rats began in the left compartment while the other half began in the right; each animal’s start side during testing was the same as its start side during preconditioning. Note, however, that start side was varied across rats to control for any possible side bias. Again, the amount of time spent in each of the two chambers and tunnel was recorded. 2.4. Drugs All drugs were prepared fresh daily. Žq.-amphetamine sulfate ŽSmithKline Beecham Pharma of Canada. was dissolved in 0.9% NaCl, and injected i.p. in a volume of 1.0 mlrkg, in a dose of 2.0 mgrkg Ž n s 20., 5 min prior to placement in the apparatus. Non-drug control animals received injections of distilled water in a volume of 1.0 mlrkg Ž n s 12. 15 min prior to the beginning of a session. SKF 82958 hydrobromide Ž"-6-chloro-7,8-dihydroxy3-allyl-1-phenyl-2,3,4,5-tetrahydro-1-phenyl-1H-3-benzazepine hydrobromide. was purchased from Research Biochemicals International, Natick, MA ŽRBI., and dissolved in a volume of 1.0 mlrkg in distilled water. Groups were injected i.p. with doses of 0.01 Ž n s 7., 0.025 Ž n s 9., 0.05 Ž n s 10., 0.075 Ž n s 12., or 0.10 Ž n s 10. mgrkg 15 min prior to placement in the apparatus, while s.c. injected animals were given doses of 0.01 Ž n s 10., 0.05 Ž n s 10., or 0.10 Ž n s 12. mgrkg 15 min prior to the beginning of a session.

113

SKF 81297 hydrobromide Ž"-6-chloro-7,8-dihydroxy1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrobromide., also purchased from RBI, was dissolved in a volume of 1.0 mlrkg in distilled water and injected i.p. in doses of 0.25 Ž n s 8., 0.50 Ž n s 13., 1.0 Ž n s 10., 2.0 Ž n s 12., or 4.0 Ž n s 5. mgrkg 15 min prior to placement in the apparatus. Gentle heating was used in preparation of the 4.0 mgrkg dose. Finally, SKF 77434 hydrobromide Ž"-7,8-dihydroxy3-allyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrochloride., purchased from RBI, was dissolved in a volume of 2.0 mlrkg in distilled water, and injected i.p. in doses of 0.20 Ž n s 8., 1.0 Ž n s 8., 5.0 Ž n s 12., or 10.0 Ž n s 7. mgrkg 15 min prior to placement in the apparatus.

3. Results Dependent variables included time spent on the drugpaired side and time spent in the tunnel on three pre-conditioning and one test days, and activity during eight conditioning sessions. Time and activity data will be treated separately. 3.1. Place conditioning For all analyses, time spent on the drug-paired side for the three pre-conditioning sessions was averaged to provide the best measure of side preference prior to conditioning. It was decided prior to behavioral testing that only data from the first test session would be used, as conditioning was expected to be strongest on this day. During pre-conditioning and test sessions, animals spent some time in the tunnel, mean Ž"S.E.M.. values ranging from 38.1 Ž"5.3. to 80.3 Ž"14.3. s per session Žsee Table 1.. Because tunnel time was observed to differ from pre-conditioning Žaveraged for three sessions. to test for some groups, F Ž1, 30. s 4.91, P - 0.05, tunnel time was removed from all sessions to eliminate any possible bias due to this variable. Thus, place conditioning was evaluated by comparing the percentage of total session time Ž900 s – tunnel time. spent in the drug-paired side during pre-conditioning and in the test. Percentage of time spent on the drug-paired side increased from pre-conditioning to test for the amphetamine but not the distilled water control group ŽFig. 1.. Groups treated with SKF 82958 s.c. during conditioning showed an increased proportion of time spent on the drug-paired side in the test with the strongest effect occurring at 0.05 mgrkg. Groups receiving this drug i.p. similarly showed an increase in percentage of time spent on the drug-paired side, except at the lowest dose, and again the strongest effect occurred at the 0.05 mgrkg dose. The effect of both SKF 81297 and SKF 77434 appeared to be systematic across doses, with the peak effect occurring in the middle

114

B.S. Abrahams et al.r European Journal of Pharmacology 343 (1998) 111–118

Fig. 1. Percentage of time spent on the drug-paired side as a function of total time spent on either side but not in the tunnel. Difference scores represent the percentage of time spent on the drug-paired side during the first test day minus the mean percentage of time spent on that side during pre-conditioning. Positive values indicate an increased amount of time on the drug-paired side during test. ) Significant increase in proportion of time in the drug-paired side in test compared to pre-conditioning.

of the dose range tested for SKF 81297, but for neither drug did any of the doses produce an effect of the same magnitude as that seen with amphetamine or the 0.05 mgrkg doses of SKF 82958 ŽFig. 1.. Statistical analyses generally confirmed this description of the results. A mixed design analysis of variance ŽANOVA. with independent groups and repeated measures on phase Žpre-conditioning versus test. was used to compare the percentage of time spent on the drug-paired side in pre-conditioning versus test for the amphetamine versus distilled water control groups. Results revealed a significant interaction, F Ž1, 30. s 4.48, p - 0.05, showing that the phase effect differed between the groups. Tests of simple effects confirmed that only the amphetamine group showed a significant increase in the proportion of time spent on the drug-paired side from pre-conditioning to test, F Ž1, 30. s 10.04, p - 0.01. Finally, planned comparisons of pre-conditioning and test for each group showed a significant effect only for the amphetamine group, t Ž19. s 3.05, p - 0.01. The ANOVA for distilled water and SKF 82958 s.c. doses yielded a significant effect of phase, F Ž1, 40. s 7.15, p - 0.05, showing that when all groups were combined animals spent a greater proportion of time on the drugpaired side in the test than in the pre-conditioning sessions. In planned comparisons of pre-conditioning and test for

each dose, only the 0.05 mgrkg s.c. dose produced a significant increase in proportion of time spent on the drug-paired side, t Ž9. s 2.26, P - 0.05. A similar mixed design ANOVA for the distilled water and SKF 82958 i.p. doses yielded no significant effects. However, as was the case for groups given SKF 82958 s.c., planned comparisons showed that only the 0.05 mgrkg i.p dose produced a significant increase in the proportion of time spent on the drug-paired side from pre-conditioning to test, t Ž9. s 2.41, P - 0.05. To directly compare the effects of those doses Ž0.01, 0.05, 0.10 mgrkg. of SKF 82958 that were given both by the s.c. and i.p. route, a three-variable ANOVA, with repeated measures on phase and independent dose and replication groups was conducted. Results revealed a significant main effects of phase, F Ž1, 53. s 6.81, P - 0.05, and a dose by phase interaction that approached significance, F Ž2, 53. s 3.03, P s 0.0565. None of the remaining main effects or interactions were significant. The lack of any significant effects involving the replication variable shows that administration of SKF 82958 by the two different routes did not affect significantly the action of the drug. As was the case for the analysis of groups receiving SKF 82958 s.c., the significant phase effect shows that when doses Žfrom both routes of administration. were combined, animals spent a greater proportion of time on

B.S. Abrahams et al.r European Journal of Pharmacology 343 (1998) 111–118

the drug-paired side in the test than in the pre-conditioning sessions. The interaction suggests that the phase effect differed depending on the dose, with doses combined over replication. Tests of simple main effects of phase at each dose yielded a significant effect for the 0.05 mgrkg dose of SKF 82958, F Ž1, 56. s 11.98, P - 0.001, revealing the source of the interaction. Neither the ANOVAs nor the planned comparisons of phase for each dose of SKF 81297 or SKF 77434 yielded any significant effects. 3.2. Locomotor actiÕity Mean number of beam breaks per min during the four drug conditioning sessions were averaged within and across sessions as were the corresponding values for vehicle sessions ŽTable 2.. As expected, amphetamine stimulated activity and distilled water did not. Generally, low doses of the dopamine D 1-like receptor agonists had little effect or slightly decreased activity whereas higher doses increased activity. However, these effects were small, none of them being of the same magnitude as the amphetamine effect. This description of the results was supported by statistical analyses. The amphetamine and distilled water groups were compared using a two variable mixed design ANOVA

Table 1 Mean Ž"S.E.M.. time Žs. spent on the drug-paired side and in the tunnel during the mean of the pre-conditioning sessions and the first test session Group

Drug paired side

Tunnel

pre-cond.

test

pre-cond.

test

434.8 Ž14.6. 426.3 Ž18.4.

498.9 Ž15.6. 413.7 Ž41.1.

46.1 Ž5.2. 48.0 Ž6.0.

52.8 Ž4.5. 57.4 Ž10.8.

SKF 82958 Žs.c.. 0.01 mgrkg 417.2 Ž19.9. 0.05 mgrkg 431.1 Ž32.1. 0.10 mgrkg 405.1 Ž28.8.

436.3 Ž46.0. 556.6 Ž54.7. 470.2 Ž46.1.

55.9 Ž6.5. 45.7 Ž4.7. 47.5 Ž5.4.

72.0 Ž11.7. 38.1 Ž5.3. 38.8 Ž7.9.

SKF 82958 Ži.p. 0.01 mgrkg 0.025 mgrkg 0.05 mgrkg 0.075 mgrkg 0.10 mgrkg

469.3 Ž38.0. 402.8 Ž50.6. 405.6 Ž25.7. 430.8 Ž30.6. 435.5 Ž35.5.

416.9 Ž39.4. 412.4 Ž44.2. 508.2 Ž46.7. 462.8 Ž32.3. 459.9 Ž49.6.

47.9 Ž6.8. 43.1 Ž6.7. 56.4 Ž6.7. 50.1 Ž7.3. 53.0 Ž7.1.

80.3 Ž14.3. 50.7 Ž5.6. 43.7 Ž4.9. 54.7 Ž7.8. 48.0 Ž2.6.

SKF 81297 0.25 mgrkg 0.50 mgrkg 1.0 mgrkg 2.0 mgrkg 4.0 mgrkg

421.9 Ž24.5. 410.5 Ž21.6. 413.7 Ž31.6. 424.6 Ž31.0. 430.8 Ž34.0.

426.5 Ž38.6. 443.1 Ž34.4. 468.4 Ž41.3. 439.6 Ž42.9. 381.4 Ž56.5.

58.2 Ž7.2. 56.7 Ž7.0. 50.1 Ž5.9. 55.4 Ž7.2. 38.6 Ž6.6.

74.5 Ž9.2. 65.3 Ž7.3. 56.1 Ž5.5. 57.3 Ž11.5. 41.4 Ž12.4.

SKF 77434 0.20 mgrkg 1.0 mgrkg 5.0 mgrkg 10.0 mgrkg

425.8 Ž15.8. 429.4 Ž29.1. 392.1 Ž20.5. 407.1 Ž34.7.

402.5 Ž59.3. 365.6 Ž35.4. 420.4 Ž34.6. 425.1 Ž63.5.

54.3 Ž7.3. 48.3 Ž7.7. 58.7 Ž7.4. 50.3 Ž8.9.

65.1 Ž10.0. 47.1 Ž9.8. 58.6 Ž6.5. 46.6 Ž9.1.

Amphetamine 2.0 mgrkg 0.0 mgrkg

115

Table 2 Beam breaks per min during drug and vehicle conditioning sessions, expressed as an average across the four conditioning days of each type Group

n

Drug Ž"S.E.M..

Vehicle Ž"S.E.M..

Amphetamine 2.0 mgrkg

20

14.28 Ž1.09. a

8.23 Ž0.85.

Distilled water 0.0 mgrkg

12

7.49 Ž0.45.

7.38 Ž0.41.

SKF 82958 Žs.c.. 0.01 mgrkg 10 0.05 mgrkg 10 0.10 mgrkg 12

6.88 Ž0.63. 10.50 Ž1.34. a 11.92 Ž1.25. a

6.95 Ž0.62. 7.29 Ž0.47. 8.04 Ž0.99.

SKF 82958 Ži.p. 0.01 mgrkg 7 0.025 mgrkg 9 0.05 mgrkg 10 0.075 mgrkg 12 0.10 mgrkg 10

7.75 Ž0.67. 7.05 Ž0.36. 7.14 Ž0.23. a 7.83 Ž0.56. a 6.70 Ž0.49.

8.07 Ž0.61. 7.51 Ž0.59. 6.17 Ž0.27. 6.36 Ž0.39. 6.25 Ž0.48.

SKF 81297 0.25 mgrkg 0.50 mgrkg 1.0 mgrkg 2.0 mgrkg 4.0 mgrkg

8 13 10 12 5

6.58 Ž0.36. 6.65 Ž0.39. 7.78 Ž0.44. a 7.88 Ž0.43. a 9.78 Ž1.16. a

6.90 Ž0.53. 6.17 Ž0.32. 6.36 Ž0.49. 6.50 Ž0.20. 7.19 Ž0.51.

SKF 77434 0.20 mgrkg 1.0 mgrkg 5.0 mgrkg 10.0 mgrkg

8 8 12 7

6.02 Ž0.47. 7.72 Ž0.40. 8.37 Ž0.75. 7.91 Ž1.09.

6.18 Ž0.63. 6.19 Ž0.29. 6.83 Ž0.46. 7.32 Ž0.26.

a

Significantly different from vehicle in tests of simple effects following a significant interaction in ANOVA.

with independent groups and repeated measures on treatment. Results revealed significant group, F Ž1, 30. s 31.08, P - 0.001, and interaction effects, F Ž1, 30. s 19.78, P 0.001. The interaction was significant because amphetamine produced a significant simple effect of treatment, F Ž1, 30. s 54.85, P - 0.001, whereas distilled water did not. Thus, amphetamine stimulated locomotor activity during conditioning sessions. A similar ANOVA for the distilled water and SKF 82958 s.c. dose groups revealed significant effects of dose, F Ž3, 40. s 5.37, P - 0.01, treatment, F Ž1, 40. s 8.63, P - 0.01, and their interaction, F Ž3, 40. s 2.85, P - 0.05. Examination of simple main effects revealed significant treatment effects for the 0.05 and 0.10 mgrkg s.c. doses of SKF 82958, F Ž1, 40. s 6.39, P - 0.05 and F Ž1, 40. s 11.21, P - 0.01, respectively, but not for the 0.01 mgrkg dose or distilled water, revealing the source of the interaction. Analysis of the distilled water and SKF 82958 i.p. dose groups revealed a significant dose by treatment interaction, F Ž5, 54. s 2.52, P - 0.05. The interaction was produced by the simple main effect of treatment for only the 0.05 and 0.075 mgrkg i.p. doses, F Ž1, 54. s 4.16, P - 0.05 and F Ž1, 54. s 12.05, P - 0.001, respectively.

116

B.S. Abrahams et al.r European Journal of Pharmacology 343 (1998) 111–118

Comparison of the doses Ž0.01, 0.05, and 0.10 mgrkg. of SKF 82958 that were given by both routes of administration suggests that the profile of effects on activity was similar for each ŽTable 2.. Thus, 0.01 mgrkg by either route produced a small decrease in activity whereas 0.05 and 0.10 mgrkg produced increases. However, the magnitude of increase produced by these higher doses was greater following s.c. administration. ANOVA followed by tests of simple main effects confirmed this description of the data. A three variable mixed design ANOVA revealed significant effects of replication, F Ž1, 53. s 11.06, P 0.01, replication by dose, F Ž2, 53. s 7.83, P - 0.01, and replication by treatment, F Ž1, 53. s 4.51, P - 0.05; a significant treatment effect, F Ž1, 53. s 8.38, P - 0.01, was also observed, but as it occurs when replications are combined and shows an effect already described above, it will not be discussed further. The significant replication effect reflects the generally higher level of activity observed in the dose groups given SKF 82958 s.c. versus those injected i.p. The replication by dose interaction reflects the fact that activity averaged over treatment was higher in the groups receiving 0.05 and 0.10 but not 0.01 mgrkg SKF 82958 by the s.c. route versus the i.p. route. The replication by treatment interaction reflects the higher magnitude of the treatment effect in the s.c. dose groups combined versus the i.p. dose groups combined. All of this is consistent with the idea that activity levels were generally higher in groups receiving SKF 82958 s.c. versus i.p. However, the lack of a replication by treatment by dose interaction is consistent with the finding that although the magnitude of activity observed in the s.c. experiment was higher, the profile of activity effects produced by SKF 82958 given by the s.c. and i.p. routes was similar. Analysis of the distilled water and SKF 81297 dose groups revealed significant dose, F Ž5, 54. s 2.98, P 0.05, treatment, F Ž1, 54. s 27.59, P - 0.001, and interaction effects, F Ž5, 54. s 5.76, P - 0.001. Simple effects analyses revealed treatment effects for the 1.0, 2.0, and 4.0 mgrkg doses, F Ž1, 9. s 11.52, P - 0.001, F Ž1, 11. s 13.15, P - 0.001, and F Ž1, 4. s 19.20, P - 0.001, respectively, but not the lower doses, revealing the source of the interaction. Analysis of the distilled water and SKF 77434 dose groups revealed a significant treatment effect, F Ž1, 42. s 8.27, P - 0.01, showing that when dose groups were combined, activity was higher on the drug conditioning than the vehicle days. Although the lack of a significant interaction precludes attributing this effect to specific doses, inspection of Table 2 reveals that doses of 1.0 and 5.0 mgrkg produced the largest effects.

4. Discussion This study is the first to report that a place preference is produced by systemic administration of a dopamine D 1-like

receptor agonist. The reliability of this result was confirmed in two separate experiments using two different routes of administrations Žs.c. and i.p... In contrast, the dopamine D 1-like receptor agonists SKF 81297 and SKF 77434 failed to produce place conditioning over a wide range of doses. Note also that all agents tested Žexcluding distilled water alone. stimulated locomotor activity at some doses, with greatest increases seen following amphetamine administration. Different routes of administration were used in our evaluation of SKF 82958 because the behavioral effects of the dopamine receptor agonist apomorphine have been shown to differ depending on route of administration ŽUngerstedt, 1979.. Thus, once we found that only a single dose of SKF 82958 given by the s.c. route produced significant place conditioning, we decided to evaluate the reliability of this effect using the i.p. route. However, the original effect was replicated at the same dose and no difference was found. The observation of a place preference following pairings of amphetamine with one side of the test apparatus is in agreement with many previous reports Žreviews: Hoffman, 1989; Carr et al., 1989.. The finding that animals treated with distilled water on both sides during conditioning showed little change in time spent on the designated side demonstrates that side preference assessed during pre-conditioning is stable over the course of conditioning sessions when no drug is given. Similarly small changes from pre-conditioning to test were seen with the lower doses of SKF 82958 and with all doses of SKF 81297 and SKF 77434. Previous studies have evaluated the place conditioning potential of the prototypical dopamine D 1-like receptor agonist SKF 38393. Results revealed a place aversion following systemic administration ŽHoffman and Beninger, 1988, 1989; White et al., 1991.. Hoffman and Beninger Ž1989. showed that the dopamine D 1-like receptor agonist fenoldopam ŽSKF 82526., which fails to cross the blood brain barrier ŽHahn et al., 1982., had no significant effect in a place-conditioning task, confirming that SKF 38393 produces its aversive effects by acting centrally. White et al. Ž1991., on the other hand, found a place preference following pairings of intra-accumbens injections of SKF 38393 with one side of a place conditioning apparatus, showing at least one site in the brain where stimulation of dopamine D 1-like receptors by SKF 38393 leads to a rewarding effect. Taken together these results suggest that SKF 38393 produces a place aversion by acting centrally at a site other than the nucleus accumbens. Results from place conditioning studies evaluating dopamine D 1-like receptor agonists show a good fit with results of self-administration studies evaluating dopamine D 1-like receptor agonists. Thus, SKF 38393 failed to support self-administration in two reports ŽWoolverton et al., 1984; Weed and Woolverton, 1995. whereas SKF 82958 was found to be self-administered by both rats ŽSelf and

B.S. Abrahams et al.r European Journal of Pharmacology 343 (1998) 111–118

Stein, 1992; Self et al., 1993, 1996. and monkeys ŽGrech et al., 1996.. SKF 81297 also was self-administered ŽWeed et al., 1993; Grech et al., 1996. whereas SKF 77434 has been reported to produce both positive ŽSelf and Stein, 1992. and negative results ŽGrech et al., 1996. in this paradigm. There appears to be a continuum of effectiveness of dopamine D 1-like receptor agonists in self-administration and place conditioning paradigms following systemic administration with the order from most to least effective being: SKF 82958G SKF 81297 ) SKF 77434 ) SKF 38393. This rank order of effectiveness of the dopamine D 1-like receptor agonists in place conditioning and self-administration experiments parallels their ability to stimulate cyclic adenosine 3X ,5X-monophosphate ŽcAMP.. Thus, SKF 82958, SKF 81297, SKF 77434, and SKF 38393 have been reported to stimulate cAMP production relative to dopamine itself, in the following respective percentage ranges: 78–148%, 68–88%, 48–55%, and 44–59% ŽAnderson et al., 1985; O’Boyle et al., 1989; Anderson and Jansen, 1990; Arnt et al., 1992; Izenwasser and Katz, 1993.. These results might suggest that the rewarding capacity of dopamine D 1-like receptor agonists is related to their efficacy at stimulating cAMP production. However, the observation by White et al. Ž1991. that intra-accumbens injections of SKF 38393 produce place preference conditioning is not consistent with this hypothesis and remains to be explained. Our observation that SKF 82958, SKF 81297, and SKF 77434 produced a mild Žrelative to amphetamine. but significant stimulation of locomotor activity is consistent with previous findings ŽMurray and Waddington, 1989; Meyer and Shults, 1993.. The fact that all agents Žexcluding distilled water alone. were able to increase locomotor activity at one or more doses, but place conditioning was seen only with 2.0 mgrkg amphetamine and 0.05 mgrkg SKF 82958 reveals a dissociation between the locomotor stimulating effect of the dopamine D 1-like receptor agonists SKF 81297 and SKF 77434 and their ability to produce place conditioning. A similar dissociation has been reported previously for adenosinergic agents ŽBrockwell and Beninger, 1996.. In conclusion, the present finding that dopamine D 1-like receptor stimulation is capable of supporting reward-related learning in the place conditioning paradigm is consistent with previous findings from self-administration studies ŽWoolverton et al., 1984; Self and Stein, 1992; Self et al., 1993, 1996; Weed et al., 1993; Grech et al., 1996.. These data add to those from studies employing other reward paradigms Že.g., Beninger and Rolfe, 1995. and receptor subtype-specific dopamine antagonists Žsee reviews by Miller et al., 1990; Beninger, 1993; Beninger and Nakonechny, 1996; Beninger and Miller, 1997. in continuing to suggest an important role for D1-like dopamine receptors in reward-related learning.

117

Acknowledgements Amphetamine was the generous gift of SmithKline Beecham Pharma Inc. of Canada. Funded by a grant to R.J.B. from the Natural Sciences and Engineering Research Council of Canada. References Anderson, P.H., Jansen, J.A., 1990. Dopamine receptor agonists: Selectivity and dopamine D1 receptor efficacy. Eur. J. Pharmacol. 188, 335–347. Anderson, P.H., Grønvald, F.C., Jansen, J.A., 1985. A comparison between dopamine-stimulated adenylate cyclase and 3H-SCH 23390 binding in rat striatum. Life Sci. 37, 1971–1983. Arnt, J., Hyttel, J., Sanchez, C., 1992. Partial and full dopamine D1-receptor agonists in mice and rats-relation between behavioural effects and stimulation of adenylate cyclase activity in vitro. Eur. J. Pharmacol. 213, 259–267. Beninger, R.J., 1993. Role of D1 and D2 receptors in learning. In: Waddington. J. ŽEd.., D1:D2 Dopamine Receptor Interactions: Neuroscience and Pharmacology. Academic Press, Amsterdam, pp. 115– 157. Beninger, R.J., Miller, R., 1997. Dopamine D1-like receptors and reward-related incentive learning. Neurosci. Biobehav. Rev., in press. Beninger, R.J., Nakonechny, P.L., 1996. Dopamine D1-like receptors and molecular mechanisms of incentive learning. In: Beninger, R.J., Palomo, T., Archer, T. ŽEds.., Dopamine Disease States. CYM Press, Madrid, pp. 407–431. Beninger, R.J., Rolfe, N.G., 1995. Dopamine D1-like receptor agonists impair responding for conditioned reward in rats. Behav. Pharmacol. 6, 785–793. Bergman, J., Kamien, J.B., Spealman, R.D., 1990. Antagonism of cocaine self-administration by selective dopamine D1 and D 2 antagonists. Behav. Pharmacol. 1, 355–364. Brockwell, N.T., Beninger, R.J., 1996. The differential role of A1 and A2 adenosine receptor subtypes in locomotor activity and place conditioning in rats. Behav. Pharmacol. 7, 373–383. Brockwell, N.T., Ferguson, D.S., Beninger, R.J., 1996. A computerized system for the simultaneous monitoring of place conditioning and locomotor activity in rats. J. Neurosci. Methods 64, 227–232. Caine, S.B., Koob, G.F., 1994. Effects of dopamine D-1 and D-2 antagonists on cocaine self-administration under different schedules of reinforcement in the rat. J. Pharmacol. Exp. Ther. 270, 209–218. Carr, G.D., Fibiger, H.S., Phillips A.G., 1989. Conditioned place preference as a measure of drug reward. In: Liebman, J.M., Cooper, S.J. ŽEds.., The Neuropharmacological Basis of Reward. Clarendon Press, Oxford, pp. 264–319. Corrigall, W.A., Coen, S.J., 1991. Cocaine self-administration is increased by both D1 and D2 dopamine antagonists. Pharmacol. Biochem. Behav. 39, 799–802. De Wit, H., Wise, R.A., 1977. Blockade of cocaine reinforcement in rats with the DA receptor blocker pimozide, but not the NA blockers phentolamine or phenoxybenyamine. Can. J. Psychol. 31, 195–203. Grech, D.M., Spealman, R.D., Bergman, J., 1996. Self-administration of D1 receptor agonists by squirrel monkeys. Psychopharmacology 125, 97–104. Hahn Jr., R.A., Wardell, J.R., Sarau, H.M., Ridley, P.T., 1982. Characterization of the peripheral and central effects of SK and F 82526, a novel dopamine receptor agonist. J. Pharmacol. Exp. Ther. 223, 305-313. Hiroi, N., White, N.M., 1991. The amphetamine conditioned place preference – Differential involvement of dopamine receptor subtypes and two dopaminergic terminal areas. Brain Res. 552, 141–152.

118

B.S. Abrahams et al.r European Journal of Pharmacology 343 (1998) 111–118

Hoffman, D.C., 1989. The use of place conditioning in studying the neuropharmacology of drug reinforcement. Brain Res. Bull. 24, 373– 387. Hoffman, D.C., Beninger, R.J., 1988. Selective D1 and D2 dopamine agonists produce opposing effects in place conditioning but not in conditioned taste aversion learning. Pharmacol. Biochem. Behav. 31, 1–8. Hoffman, D.C., Beninger, R.J., 1989. The effects of selective dopamine D1 and D2 receptor antagonists on the establishment of agonist-induced place conditioning in rats. Pharmacol. Biochem. Behav. 33, 273–279. Hoffman, D.C., Dickson, P.R., Beninger, R.J., 1988. The dopamine D2 receptor agonists, quinpirole and bromocriptine produce conditioned place preferences. Prog. Neuro-Psychopharmacol. Biol. Psychiatr. 12, 315–322. Izenwasser, S., Katz, J.L., 1993. Differential efficacies of dopamine D1 receptor agonists for stimulating adenylyl cyclase in squirrel monkey and rat. Eur. J. Pharmacol. 246, 39–44. Kebabian, J.W., Calne, D.B., 1979. Multiple receptors for dopamine. Nature 277, 93–96. Koob, G.F., Le, H.T., Creese, I., 1987. The D1 dopamine receptor antagonist SCH 23390 increases cocaine self-administration in the rat. Neurosci. Lett. 79, 315–320. Mackey, W.B., van der Kooy, D., 1985. Neuroleptics block the positive reinforcing effects of amphetamine but not of morphine as measured by place conditioning. Pharmacol. Biochem. Behav. 22, 101–106. Meyer, M.E., Shults, J.M., 1993. Dopamine D1 receptor family agonists, SK and F77434, and SK and F82958, differentially affect locomotor activities in rats. Pharmacol. Biochem. Behav. 46, 269–274. Miller, R., Wickens, J.R., Beninger, R.J., 1990. Dopamine D-1 and D-2 receptors in relation to reward and performance: A case for the D-1 receptor as a primary site of therapeutic action of neuroleptic drugs. Prog. Neurobiol. 34, 143–183. Mithani, S., Martin-Iverson, M.T., Phillips, A.G., Fibiger, H.C., 1986. The effects of haloperidol on amphetamine- and methylphenidate-induced conditioned place preferences and locomotor activity. Psychopharmacology 90, 247–252. Morency, M.A., Beninger, R.J., 1986. Dopaminergic substrates of cocaine-induced place conditioning. Brain Res. 399, 33–41. Murray, A.M., Waddington, J.L., 1989. The induction of grooming and vacuous chewing by a series of selective D-1 dopamine receptor agonists: Two directions of D-1:D-2 interaction. Eur. J. Pharmacol. 160, 377–384. O’Boyle, K.M., Gaitanopoulos, D.E., Brenner, M., Waddington, J.L., 1989. Agonist and antagonist properties of benzazepines and thienopyridine derivatives at the D1 dopamine receptor. Neuropharmacology 28, 401–404. Risner, J.E., Jones, B.E., 1976. Role of noradrenergic and dopaminergic processes in amphetamine self-administration. Pharmacol. Biochem. Behav. 5, 477–482.

Roberts, D.C.S., Vickers, G., 1984. Atypical neuroleptics increase selfadministration of cocaine: An evaluation of a behavioural screen for antipsychotic activity. Psychopharmacology 82, 135–137. Roberts, D.C.S., Vickers, G., 1987. The effect of haloperidol on cocaine self-administration is augmented with repeated administrations. Psychopharmacology 93, 526–528. Seeman, P., Van Tol, H.H.M., 1994. Dopamine receptor pharmacology. Trends Pharmacol. Sci. 15, 264–270. Self, D.W., Stein, L., 1992. The D1 agonists SKF 82958 and SKF 77434 are self administered by rats. Brain Res. 582, 349–352. Self, D.W., Lam, D.M., Kossuth, S.R., Stein, L., 1993. Effects of D1 and D2-selective antagonists on self-administration of the D1 agonist SKF 82958. In: Harris, L. ŽEds.., NIDA Research Monograph 132, College on the Problems of Drug Dependence. NIDA, Washington, DC, p. 230. Self, D.W., Belluzzi, J.D., Kossuth, S., Stein, L., 1996. Self-administration of the D1 agonist SKF 82958 is mediated by D1, not D2, receptors. Psychopharmacology 123, 303–306. Spyraki, C., Fibiger, H.C., Phillips, A.G., 1982. Attenuation by haloperidol of place preference conditioning using food reinforcement. Psychopharmacology 77, 379–382. Ungerstedt, U., 1979. Central dopamine mechanisms of unconditioned behaviour. In: Horn, A.S., Korf, J., Westerink, B.H.C. ŽEds.., The Neurobiology of Dopamine. Academic Press, London, pp. 577–596. Weed, M.R., Vanover, K.E., Woolverton, W.L., 1993. Reinforcing effects of the D1 dopamine agonist SKF 81297 in rhesus monkeys. Psychopharmacology 113, 51–52. Weed, M.R., Woolverton, W.L., 1995. The reinforcing effects of dopamine D1 receptor agonists in rhesus monkeys. J. Pharmacol. Exp. Ther. 275, 1367–1374. White, N.M., Packard, M.G., Hiroi, N., 1991. Place conditioning with D1 and D2 agonists injected peripherally or into the nucleus accumbens. Psychopharmacology 103, 271–276. Wise, R.A., Murray, A., Bozarth, M.A., 1990. Bromocriptine self-administration and bromocriptine-reinstatement of cocaine-trained and heroin-trained lever pressing in rats. Psychopharmacology 100, 355– 361. Willner, P., Scheel-Kruger, J. ŽEds.., 1991. The Mesolimbic Dopamine ¨ System: From Motivation to Action. John Wiley and Sons, Chichester. Woolverton, W.L., Goldberg, L.I., Ginos, J.Z., 1984. Intravenous self-administration of dopamine receptor agonists by rhesus monkeys. J. Pharmacol. Exp. Ther. 230, 678–683. Yokel, R.A., Wise, R.A., 1975. Increased lever pressing for amphetamine after pimozide in rats: Implications for a dopamine theory of reward. Science 187, 547–549. Yokel, R.A., Wise, R.A., 1976. Attenuation of intravenous amphetamine reinforcement by central DA blockade in rats. Psychopharmacology 48, 311–318.